Determination of PCB

PCB were determined by a modified EPA Method 1668C (2010). Briefly, 25 g of soil before and after washing by either water or HA solutions were spiked with 50 μL of 2,4,5,6-tetrachloro-m-xylene solution at a concentration of 10 mg L⁻¹, added with 50 ml of n-hexane and submitted to PCB extraction in a ultrasonication bath (SONICA, 3200M S3, of Soltec) for 30 min. After sonication, the suspension was cooled at room temperature and the organic phase filtered through Whatman 42 filter paper containing sodium sulfate at the top of the filter in order to remove residual water. The extraction was repeated twice, and the organic extracts were combined, and rotoevaporated to a final volume of 10 mL. A further purification was achieved by loading 2 mL of the organic extract through a Bond Elut Florisil cartridge (1 g/6 mL, Agilent Technologies), previously conditioned with 10 ml of n-hexane, and, then, eluted with 10 mL of fresh n-hexane. This volume was further concentrated to 2 mL and added, before GC-MS analysis, with 10 μL of a quintozene (Supelco) 1 mg mL⁻¹ solution in hexane as internal standard.

The determination of PCB was accomplished by a GC-MS system consisting of an Agilent 7890 gas-chromatograph, equipped with a split/splitless injector, a HP-5MS capillary column (30 m × 0.25 mm, Agilent Tech., USA), and an Agilent 5975B mass spectrometry detector. The experimental conditions for GC analyses were the following: (1) initial temperature of 80 °C, hold time 0 min.; (2) rate of 3 °C min⁻¹ up to 250 °C, hold time 0 min; (3) rate of 3 °C min⁻¹ up to 300 °C, hold time 2 min. The total GC run time was 63.66 min. Helium was the carrier gas at 1.0 mL min⁻¹ and the splitless-flow was used. The inlet-line temperature of the GC was set at 250 °C, MS source at 150 °C and the mass transfer line at 300 °C. A solvent delay time of 5 min was applied before spectra acquisition to reduce filament consumption. Mass spectrometer operated in electron impact ionization (EI)/selected ion monitoring (SIM) mode. Three PCB standard solutions (dr. Ehrenstorfer GmBh) were used for qualitative and quantitative analysis, at 10 ng μl⁻¹ in iso-octane containing the following PCB congeners: 1. 77, 81, 105, 114, 118, 123, 126, 156, 157, 167, 169, and 189; 2. 28, 52, 95, 99, 101, 105, 110, 118, 138, 146, 149, 151, 153, 170, 177, 180, 183, and 187; 3. PCB 209. The 2,4,5,6-tetrachloro-m-xylene (CPAchem Ltd) at 100 ng μl⁻¹ in methanol and quintozene were used as spike and internal standard, respectively. Qualitative analysis was performed by comparing retention time and m/z of PCB congeners occurring in the certified standard mixtures with GC-MS peaks, while quantitative analysis was achieved by using the GC response factor of each target PCB obtained with five-point calibration curves of the same certified mixture standard. Each sample was injected twice and average and standard deviation were calculated.